Working with Data SCIENTIFIC VISUALIZATION UNIT 5.00 SYNTHESIZE DATA FOR SCIENTIFIC & TECHNICAL VISUALIZATIONS

Evaluate methods for displaying data using charts and graphs OBJECTIVE 5.01

Purpose  Charts and graphs are used to create a physical representation on the numbers in a dataset  Graphic representations of numerical data are:  Encourage quick comparison of different sets of data  Reveals several layers of detail that would be missed as raw data  Provides a clearer message than huge sets of numbers  Easier to remember

Common Chart Components  Title – Short overview of the information  Axis – Reference lines in a coordinate system  X-axis – horizontal  Y-axis – vertical  Label – Common data categories  Legend – List of symbols used in the chart  Data Value – Text showing actual value of an item  Callouts – Descriptive text in charts which is not data

Common Chart Components  Frame  Defines the plot area boundaries  Grid  Scale guideline  Tick  Shorter line than grid to indicate a scale division  Scale  Evenly spaced increments for measuring the size or position of plotted data values

Selecting the Appropriate Chart  Different chart types are used depending on the question you are asking of the data  Three Main Chart Types:  Bar/Column – Used for comparing discrete, unrelated categories of data  Pie – Used for showing parts of a whole or percentages  Line – Used for comparing related variables and relationships over time

Bar/Column Charts  Each bar/column represents a data element  Commonly consists of:  Linear scale along the vertical, Y- axis  Category scale along the horizontal, X-axis Variations:  Stacked Column:  Shows several sets of related data that add up to a whole in columns  Grouped Columns  Columns slightly overlap with the tallest column in the back of the set  Histogram  Displays frequencies or relative frequencies using columns that touch  Used to determine symmetrical distributions

Pie Charts  Each segment is angled at the same percent as the data it represents  Data needs to be converted to percentages  Combined wedges always result in 100% when added together  Data may be contiguous or simultaneous in time Variations:  Exploded Pie Chart  Used to highlight one slice for emphasis  3D Pie Chart  Used for aesthetic purposes  Often makes the data difficult to read

Line Charts  Each line represents the changes in a variable over a distinct time/temperature range  Provides a clean, accurate picture of growth Variations:  Step Line Chart  Plots abrupt changes rather than gradual using plateaus of time values  Area Chart  Shades in the area between the X- axis and line  Scatterplot  Representation between two variables where the data points are not connected  Uses a line of best fit (average) to show relationships/trends

Other Ways to Represent Data  Not all data can or should be represented using a chart  Other common data visualizations include:  Venn Diagram – Used to show relationships between sets of data (similarities vs. differences)  Flowchart – Visual representation of a process used to solve problems using specific set of symbols to represent  Infographic – Combination of text, graphics and data to create a visually impactful message Symbol Chart:

Describe the steps of a design brief OBJECTIVE 5.02

Purpose of Design Briefs  Design briefs are documents used to plan out a project  Different industries and companies have different requirements for their design briefs but they all have four components in common:  Identify the main problem to solve  Describe a plan for solving that problem  Give a timeframe for when the project will be completed  Report on the results after completing the project  Common Examples:  Science – Research Proposal or Research Plan  Business – Business Plan  Game Design – Game Design Document (GDD)

Using the Engineering Design Process  Ask Questions  Identify the problem that needs to be solved  Conduct a literature review  Form a Hypothesis  Determine a solution to the question  Plan & Design  Put your ideas on paper, gather necessary resources to test your hypothesis  Experiment & Collect Data  Conduct your experiment  Evaluate Your Results  Analyze the data  Determine how you will present your results to others: written report, slideshow presentation, etc. Engineering Design Process: Robert Bourgeois, Durham School of the Arts

Selecting a Topic Brainstorming  Many projects begin by brainstorming during their asking phase  Brainstorming is:  A process that identifies as many possible answers to the problem being examined  All ideas are valuable – If this session is a group effort, these ideas are not criticized or discussed until the session is finished  Typically has a moderator to make sure everyone’s ideas are heard and the process is orderly Nominal Group Technique  While there are many ways to determine which idea to pursue, one popular way involves nominal group technique  A way of organize information to show relative importance based on ranking by individuals on the team  Typically, each item is assigned a numerical value based on individual ranks  These are used to reduce the number of items in the list based on a top set of values (i.e. – keep top 5)  The process is repeated until there is a single idea remaining

Storyboarding: Visual Planning  Some industries use storyboarding as part of the planning process  Animation  Film  Game Design  Storyboarding is:  A series of quick sketches in sequence that illustrate the main ideas of the project  Generally not high detailed or quality drawings  Should include text, audio and video techniques, and direction for things like camera angles or specific actions necessary for a scene

Form a Hypothesis  An educated guess about how things work  Often written like this: "If _____[I do this] _____, then _____[this]_____ will happen."  Should be something that you can actually test  The main idea of your hypothesis is to state the relationship between the dependent and independent variables  The most powerful results come from controlled experiments where units being measured are grouped to keep unwanted errors from affecting results

Experimental Variables Independent Variables  What you believe might influence the outcome measure  Manipulated by the research team  Ex. – Light colors  Plotted on X-axis Dependent Variables  Modified based on independent variable  Prediction being made  Ex. – Growth of plants  Plotted on Y-axis

Different Design Types Data Driven Design  Collects data as numerical values that can be compared with each other  Examples – graphs, charts and tables Concept Driven Design  Used to explain a concept, idea or theory  Examples – how a car works, the water cycle, how to draw a tree

Conduct Experiment  Gather all the resources needed to conduct your experiment based on your design  Equipment  Dependent and independent variable sources  Special Personnel  Follow your research plan and collect data

Reporting on Results Written Report  Different industries/companies have different requirements  Science: Research Report  Often uses American Psychological Association (APA) writing conventions  Each section distinct and fully described  Game Design: GDD  States the core idea of the game  Identifies several concrete ideas that are critical to the game Multimedia Slideshow  Determine the target audience  Make sure the wording/material is appropriate  Use a template for consistency  Edit Master Slide to meet your needs  Edit the contents including graphics, videos, charts, etc.  General Rule - try to stick to short statements  7 lines of text, 7 words per line

Methods for Creating Presentations IDEAL Method  I:  Identify the problem  D:  Define a plan of work  E:  Explore the problem through research  A:  Act on the problem’s solution  L:  Look back at the process SAFE Method  S: Simple  If you keep the presentation simple, it will save you time, money, confusion and presentation problems  A: Appropriate  Make sure the content is appropriate for the audience  F: Functional  If a visual confuses an audience more than it helps to understand the topic, then it is not performing its function  E: Economical  Concentrate on the important items in your presentation so it only contains the elements needed to convey your message in a short period of time

Interpret data for use in charts and graphs OBJECTIVE 5.03

Data Collection Empirically Derived  Physically measured  Examples – Using a ruler to measure length, height, width  This kind of data does not require any math to collect Computationally Derived  Requires use of a mathematical formula to determine result  Examples – Calculate the area/volume/circumference

Measuring Data Scalar Measurements  Data explaining the magnitude of the variable being measured  Lacks measurement of direction  Common scalar measurements:  Time (years, months, days, hours, etc.)  Volume (milliliters, liters, etc.)  Weight (ounces, grams, etc.)  Speed  Temperature Vector Measurements  Data that contains both magnitude and direction of the variable being measured  Described using mathematical formulas  Common vector measurements  Increases/decreases in temperature  Velocity (rate an object changes position)  Directional measurements applied to scalar quantities

Basic Data Types Parametric  Used for predictable distributions  Interval  Measurements where the distance between values is the same  Example – Changes in temperature  Ratio  Values compared as multiples of one another  Value of 0 has meaning!  Example – Comparing height of people at different ages Non-parametric  Does not assume any predictable distribution  Nominal  Categorical where order is arbitrary  Example – Ethnicity  Ordinal  Categorized into logical order  Example – Likert Scales

Quantitative vs. Qualitative Data Quantitative  Deals with numbers  Quantities  Can be measured  Length, height, weight, temperature, distance, etc. Qualitative  Deals with descriptions  Quality  Can be observed but not measured  Color, texture, smell, interviews, etc.

Common Quantitative Calculations  Mean – Arithmetic Average  To calculate – Add all the given values and divide by the total number of values  Median – Middle  Defined as the middle value of several readings where all values are placed in an increasing or decreasing order  Mode – Most Common  Defined as the most commonly occurring value found in a group of consistent readings /

Identify Quantitative vs. Qualitative Data in the Image Below

Quantitative vs. Qualitative Data Quantitative  Image or frame size  Measured as width x height in inches or centimeters  Weight  Measured in ounces, pounds or grams  Value  Measured in Dollars or Euros  Number of people, buildings, plants Qualitative  Shades of colors used  How the picture smells  Texture of the brush strokes  Feeling given by the view presented  Stroke quality  Master or amateur?

Working with Cartesian Coordinate Systems  A way of working with and defining positions in space  Maps, computer monitors, televisions and 3D modeling all operate using Cartesian coordinates  Plotting data points in 2D  X and Y coordinates  X – Horizontal  Y – Vertical  Each quadrant identified numerically  Positive values move up and right  Negative values move down and left  Plotting data in 3D  Adds Z coordinate  Provides depth  Origin is typically (0, 0, 0)

Regression Line  A line drawn through a graph of 2 variables  The line is chosen so it comes as close as possible to the majority of the data points without specifically connecting points  Used to show trends in data  The closer the points are to the line, the stronger the relationship  Direct or Positive Relationship  Points move from the lower left to upper right  An increase in X is more likely associated with an increase in Y  Indirect or Negative Relationship  Points move from upper left to lower right  And increase in X is more likely associated with a decrease in Y

Data Relationships Direct/Positive RelationshipIndirect/Negative Relationship

Apply data to make an appropriate graph OBJECTIVE 5.04

Getting Started  Are you creating the chart manually?  Gather necessary resources: paper, pencil, ruler  Are you creating the chart digitally?  Determine spreadsheet application you will use  Enter the data into the spreadsheet  Next, determine the appropriate type of chart to use  Bar/Column – Comparisons  Pie – Percentages  Line – Trends over time  Use the spreadsheets chart wizard to create your chart  Edit the information as needed (enter title, labels, scale, etc.)

Entering Data in a Spreadsheet  Spreadsheets are made up of cells:  Columns are identified by letters  Rows are identified by numbers  Combined, they form a cell reference – A1  Label  Alphanumeric text or numbers not used in calculations  Values  Numbers used for calculations  Formulas  Statement used to perform a calculation - =A1+B1  Function  Formulas that are built into the program for common calculations

Manipulating Data Common FunctionsOrder of Operations  PEMDAS  Parenthesis  Exponents  Multiplication  Division  Addition  Subtraction  SUM  Adds a range of values  Average  Calculates the range’s mean  Min  Finds the smallest value in a range  Max  Finds the largest value in a range  If  Performs a logical test

Steps to Making Charts Using a Spreadsheet  Once the data is entered, determine what kind of chart is needed to answer the question you are asking  Bar/Column – Compare datasets  Pie – Percentages or parts of a whole  Line – Changes over time or temperature  Select the data you want to create a chart about  Make sure if you need to perform any calculations, you have already done so before making a chart  Use your application’s chart wizard  The wizard will do most of the work for you based on your selections  However, some alterations may be needed and you might need to spend some time getting accustomed to the application you intend to use before making any charts

Rules for Creating Charts  One slide – One message – One chart  Each chart should have a single, well defined message based on the data and question asked of it  Keep graphs simple – Let the data do the talking  Avoid 3D, extra colors, and pictures  Use graphing paper or a spreadsheet application  Determine the correct chart type based on the question being asked of the data  Use the largest values to determine each axis scale and fill most of the chart area with your visualization  Each graphed item represents:  Category – name of measurable item and identified with a label  Data Point – Quantity associated  Plot the independent/control variable on the x-axis  Plot the dependent variable on the y-axis  Label each axis and provide regular numerical scale values  Most charts start at an origin of (0,0)  Provide a descriptive title  Use a legend to clarify datasets (especially for line graphs)